Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3061 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0785. Min Act: -0.0001

Data Index: 782138 (C4 (Web Text))

Max Activating Token Index: 433

Text #1

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0765. Min Act: -0.0001

Data Index: 313124 (C4 (Web Text))

Max Activating Token Index: 372

Text #2

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0699. Min Act: -0.0001

Data Index: 426866 (C4 (Web Text))

Max Activating Token Index: 605

Text #3

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0628. Min Act: -0.0000

Data Index: 162674 (C4 (Web Text))

Max Activating Token Index: 356

Text #4

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0652. Min Act: -0.0001

Data Index: 939350 (C4 (Web Text))

Max Activating Token Index: 507

Text #5

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0611. Min Act: -0.0001

Data Index: 536683 (C4 (Web Text))

Max Activating Token Index: 253

Text #6

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0673. Min Act: -0.0001

Data Index: 1594503 (Python Code)

Max Activating Token Index: 678

Text #7

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0549. Min Act: -0.0001

Data Index: 1121797 (C4 (Web Text))

Max Activating Token Index: 1014

Text #8

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0571. Min Act: -0.0000

Data Index: 1301701 (C4 (Web Text))

Max Activating Token Index: 832

Text #9

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0566. Min Act: 0.0000

Data Index: 1199784 (C4 (Web Text))

Max Activating Token Index: 534

Text #10

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0549. Min Act: -0.0001

Data Index: 1201530 (C4 (Web Text))

Max Activating Token Index: 398

Text #11

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0595. Min Act: -0.0001

Data Index: 773559 (C4 (Web Text))

Max Activating Token Index: 627

Text #12

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0542. Min Act: -0.0000

Data Index: 983733 (C4 (Web Text))

Max Activating Token Index: 175

Text #13

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0504. Min Act: -0.0000

Data Index: 1234285 (C4 (Web Text))

Max Activating Token Index: 821

Text #14

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0533. Min Act: -0.0001

Data Index: 1621854 (Python Code)

Max Activating Token Index: 533

Text #15

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0509. Min Act: -0.0000

Data Index: 1531558 (Python Code)

Max Activating Token Index: 603

Text #16

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0400. Min Act: -0.0001

Data Index: 465952 (C4 (Web Text))

Max Activating Token Index: 575

Text #17

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0487. Min Act: -0.0000

Data Index: 311104 (C4 (Web Text))

Max Activating Token Index: 742

Text #18

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0522. Min Act: -0.0000

Data Index: 260396 (C4 (Web Text))

Max Activating Token Index: 480

Text #19

Max Range: 0.0785. Min Range: -0.0785

Max Act: 0.0496. Min Act: -0.0000

Data Index: 978504 (C4 (Web Text))

Max Activating Token Index: 360